CN116861204B - Intelligent manufacturing equipment data management system based on digital twinning - Google Patents
Intelligent manufacturing equipment data management system based on digital twinning Download PDFInfo
- Publication number
- CN116861204B CN116861204B CN202311134407.9A CN202311134407A CN116861204B CN 116861204 B CN116861204 B CN 116861204B CN 202311134407 A CN202311134407 A CN 202311134407A CN 116861204 B CN116861204 B CN 116861204B
- Authority
- CN
- China
- Prior art keywords
- target
- sampling period
- decision
- target sampling
- sampling
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000004519 manufacturing process Methods 0.000 title claims abstract description 91
- 238000013523 data management Methods 0.000 title claims abstract description 21
- 238000005070 sampling Methods 0.000 claims abstract description 386
- 238000003066 decision tree Methods 0.000 claims abstract description 100
- 238000001514 detection method Methods 0.000 claims abstract description 73
- 238000012545 processing Methods 0.000 claims abstract description 19
- 238000004364 calculation method Methods 0.000 claims description 24
- 238000004422 calculation algorithm Methods 0.000 claims description 23
- 239000011159 matrix material Substances 0.000 claims description 17
- 238000012790 confirmation Methods 0.000 claims description 16
- 238000010276 construction Methods 0.000 claims description 12
- 238000006243 chemical reaction Methods 0.000 claims description 8
- 238000007405 data analysis Methods 0.000 claims description 8
- 238000010606 normalization Methods 0.000 claims description 4
- 238000012163 sequencing technique Methods 0.000 claims description 3
- 239000013598 vector Substances 0.000 claims description 3
- 238000000034 method Methods 0.000 description 30
- 238000010586 diagram Methods 0.000 description 12
- 230000008569 process Effects 0.000 description 11
- 238000007637 random forest analysis Methods 0.000 description 8
- 238000012549 training Methods 0.000 description 8
- 230000002159 abnormal effect Effects 0.000 description 7
- 238000004458 analytical method Methods 0.000 description 7
- 238000005516 engineering process Methods 0.000 description 6
- 230000006870 function Effects 0.000 description 6
- 230000010354 integration Effects 0.000 description 5
- 230000009467 reduction Effects 0.000 description 4
- 238000004140 cleaning Methods 0.000 description 3
- 238000005520 cutting process Methods 0.000 description 3
- 238000005265 energy consumption Methods 0.000 description 3
- 238000012544 monitoring process Methods 0.000 description 3
- 238000005457 optimization Methods 0.000 description 3
- 238000013450 outlier detection Methods 0.000 description 3
- 238000007781 pre-processing Methods 0.000 description 3
- 238000003672 processing method Methods 0.000 description 3
- 238000013138 pruning Methods 0.000 description 3
- 238000003860 storage Methods 0.000 description 3
- 238000013480 data collection Methods 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 238000003745 diagnosis Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000003754 machining Methods 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 238000003908 quality control method Methods 0.000 description 2
- 230000002829 reductive effect Effects 0.000 description 2
- 230000002441 reversible effect Effects 0.000 description 2
- 206010063385 Intellectualisation Diseases 0.000 description 1
- 230000002411 adverse Effects 0.000 description 1
- 238000004220 aggregation Methods 0.000 description 1
- 230000002776 aggregation Effects 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 238000013145 classification model Methods 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000007906 compression Methods 0.000 description 1
- 230000006835 compression Effects 0.000 description 1
- 238000004590 computer program Methods 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 238000012217 deletion Methods 0.000 description 1
- 230000037430 deletion Effects 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 238000009499 grossing Methods 0.000 description 1
- 238000003780 insertion Methods 0.000 description 1
- 230000037431 insertion Effects 0.000 description 1
- 230000000670 limiting effect Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 230000006855 networking Effects 0.000 description 1
- 238000013439 planning Methods 0.000 description 1
- 238000010187 selection method Methods 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 238000000638 solvent extraction Methods 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/10—Pre-processing; Data cleansing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/243—Classification techniques relating to the number of classes
- G06F18/24323—Tree-organised classifiers
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/10—Office automation; Time management
- G06Q10/103—Workflow collaboration or project management
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/04—Manufacturing
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Human Resources & Organizations (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Strategic Management (AREA)
- Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- Marketing (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Manufacturing & Machinery (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Testing And Monitoring For Control Systems (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The application relates to the technical field of digital data processing, in particular to a digital twinning-based intelligent manufacturing equipment data management system, which comprises: the acquisition module is used for confirming a preset number of sampling samples based on the operation data; the confirming module is used for confirming sample decision similarity coefficients based on randomly calling a preset number of sampling samples and sampling samples of other sampling periods before a target sampling period; the computing module is used for confirming feature decision similarity coefficients according to the preset number of attribute features of the sampling samples randomly fetched in the target sampling period; the judging module is used for judging whether the current decision tree meets the establishment condition or not based on the sample decision similarity coefficient and the characteristic decision similarity coefficient; the detection module is used for constructing a state detection model according to the decision trees with preset quantity so as to confirm the running state of the intelligent manufacturing equipment through the state detection model, thereby improving the detection precision and further reducing the detection working cost.
Description
Technical Field
The application relates to the technical field of digital data processing, in particular to a digital twinning-based intelligent manufacturing equipment data management system.
Background
Digital twinning is a technique that combines a physical entity with its digitized model. The method constructs a digital model corresponding to the physical entity through collecting, integrating and analyzing the data of the physical entity so as to realize simulation, monitoring and optimization of the physical entity. The core of the digital twin technology is to correspond and synchronize the data of the physical entity with the digital model. Through technical means such as sensors, the internet of things and the like, various data of physical entities such as temperature, pressure, vibration, position and the like can be collected and monitored in real time. These data can be used to construct a digitized model and update synchronously with the model so that the digitized model can accurately reflect the state and behavior of the physical entity.
The traditional method for detecting the operation state of the intelligent manufacturing equipment based on the digital twin technology is to establish a corresponding digital twin model for the data collected by the equipment, convert the operation data of the equipment into a digital form and realize the analysis of the operation state of the intelligent manufacturing equipment. The traditional digital twin technology-based detection of the running state of the intelligent manufacturing equipment does not optimize the classification model, so that the classification effect is poor, the detection precision of the running state of the intelligent manufacturing equipment is low, and the detection working cost is high.
Disclosure of Invention
In view of the foregoing, it is necessary to provide a digital twin-based data management system for an intelligent manufacturing apparatus, which improves the detection accuracy and further reduces the detection cost compared with the conventional detection system for the operation state of the intelligent manufacturing apparatus based on the digital twin technology.
The first aspect of the present application provides a digital twinning-based intelligent manufacturing equipment data management system, which is applied to the field of digital data processing, and the system comprises: the intelligent manufacturing equipment comprises an acquisition module, a data acquisition module and a data analysis module, wherein the acquisition module is used for acquiring operation data of the intelligent manufacturing equipment and confirming a preset number of sampling samples based on the operation data, and the sampling samples comprise attribute characteristics of the preset number; the confirming module is used for confirming sample decision similarity coefficients corresponding to the target sampling period based on randomly calling a preset number of sampling samples and sampling samples of other sampling periods before the target sampling period in the target sampling period, wherein one sampling period corresponds to one decision tree; the computing module is used for confirming feature decision similarity coefficients corresponding to the target sampling period according to the preset number of attribute features of the sampling samples randomly fetched in the target sampling period; the judging module is used for judging whether a decision tree corresponding to the current target sampling period meets the establishment condition or not based on the sample decision similarity coefficient and the characteristic decision similarity coefficient so as to enter the condition of the next sampling period; the detection module is used for constructing a state detection model according to decision trees corresponding to a preset number of sampling periods so as to confirm the running state of the intelligent manufacturing equipment through the state detection model.
In one embodiment, the confirmation module is configured to confirm, in a target sampling period, a sample decision similarity coefficient corresponding to the target sampling period based on randomly calling a preset number of sampling samples and sampling samples of other sampling periods before the target sampling period, where one sampling period corresponds to one decision tree, and specifically includes: the first sub-confirmation module is used for randomly calling a preset number of sampling samples in a target sampling period and carrying out size sorting on the repetition frequency of the sampling samples in other sampling periods before the target sampling period to construct a target sampling sequence corresponding to the target sampling period; and the second sub-confirmation module is used for calculating sample decision similarity coefficients corresponding to the target sampling period based on the maximum repetition frequency of the target sampling sequence, the number of the sampling samples and the attribute feature sequences corresponding to each sampling sample.
In one embodiment, the second sub-confirmation module is configured to calculate, based on a maximum repetition frequency of the target sampling sequence, the number of sampling samples, and an attribute feature sequence corresponding to each sampling sample, a sample decision similarity coefficient corresponding to a target sampling period, and specifically includes:
Wherein,is the firstSample decision similarity coefficients corresponding to the individual target sampling periods,is the firstThe maximum repetition frequency of the target sample sequence corresponding to the target sample period,is the firstThe number of sample samples of the target sample sequence corresponding to the target sample period,is the firstThe first target sampling sequence corresponding to each target sampling periodThe sequence of attribute features corresponding to the individual sample samples,is the firstThe first target sampling sequence corresponding to each target sampling periodThe sequence of attribute features corresponding to the individual sample samples,to calculate the edit distance between two attribute feature sequences.
In one embodiment, the calculating module is configured to determine, according to a preset number of attribute features of the sampling samples randomly fetched in the target sampling period, a feature decision similarity coefficient corresponding to the target sampling period, and specifically includes: the first sub-calculation module is used for confirming a sampling attribute feature sequence corresponding to the target sampling period according to the attribute features of the preset number of sampling samples randomly fetched in the target sampling period; the second sub-calculation module is used for constructing a target attribute feature matrix corresponding to the target sampling period based on the target attribute feature sequence corresponding to the target sampling period and other attribute feature sequences corresponding to other sampling periods before the target sampling period; the third sub-calculation module is used for carrying out preset conversion algorithm processing on the target attribute feature matrix and confirming a target attribute feature sequence corresponding to a target sampling period; and the fourth sub-calculation module is used for calculating the feature decision similarity coefficient corresponding to the target sampling period based on the number of data points in the target attribute feature sequence and the data value corresponding to each data point.
In one embodiment, the second sub-calculation module is configured to construct a target attribute feature matrix corresponding to the target sampling period based on the target attribute feature sequence corresponding to the target sampling period and other attribute feature sequences corresponding to other sampling periods before the target sampling period, and specifically includes: the construction unit is used for sequencing other attribute feature sequences corresponding to other sampling periods before the target sampling period according to the columns, and constructing other attribute feature matrixes corresponding to the other sampling periods; and the confirmation unit is used for carrying out comparison processing of preset rules through the column vectors of the target attribute feature sequences corresponding to the target sampling periods and other attribute feature matrices corresponding to other sampling periods, and confirming the target attribute feature matrices corresponding to the target sampling periods.
In one embodiment, the fourth sub-calculating module is configured to calculate, based on the number of data points in the target attribute feature sequence and a data value corresponding to each data point, a feature decision similarity coefficient corresponding to a target sampling period, and specifically includes:
wherein,is the firstThe feature decision similarity coefficients corresponding to the individual target sampling periods, Is the firstThe number of data points in the target attribute feature sequence corresponding to the target sampling period,is the firstThe first target attribute feature sequence corresponding to each target sampling periodData values for data points.
In one embodiment, the determining module is configured to determine, based on the sample decision similarity coefficient and the feature decision similarity coefficient, whether a decision tree corresponding to a current target sampling period meets an establishment condition, so as to enter a next sampling period condition, and specifically includes: the first sub-judging module is used for inputting the sample decision similarity coefficient and the characteristic decision similarity coefficient into a preset decision similarity calculation formula and calculating the decision similarity of a decision tree corresponding to the target sampling period; the second sub-judging module is used for determining that the target sampling period needs to be resampled when the decision similarity of the decision tree corresponding to the target sampling period is larger than a preset similarity threshold; and the third sub-judging module is used for determining that the decision tree corresponding to the target sampling period meets the condition when the decision similarity of the decision tree corresponding to the target sampling period is smaller than or equal to a preset similarity threshold value, and entering the condition of the next sampling period.
In one embodiment, the first sub-judging module is configured to input the sample decision similarity coefficient and the feature decision similarity coefficient into a preset decision similarity calculation formula, and calculate the decision similarity of the decision tree corresponding to the target sampling period, and specifically includes:
wherein,is the firstSample decision similarity coefficients corresponding to the individual target sampling periods,is the firstSample decision similarity coefficients corresponding to the individual target sampling periods,is a normalization function.
In one embodiment, the detection module is configured to construct a state detection model according to a decision tree corresponding to a preset number of sampling periods, so as to confirm, by using the state detection model, an operation state of the intelligent manufacturing apparatus, and specifically includes: the first sub-detection module is used for constructing a state detection model according to decision trees corresponding to a preset number of sampling periods; the second sub-detection module is used for inputting real-time operation data acquired in real time and used for acquiring the intelligent manufacturing equipment into the state detection model and confirming the operation state of the intelligent manufacturing equipment, wherein the operation state comprises a fault state, an early warning state and a normal state.
In one embodiment, the second sub-detection module is configured to input real-time operation data acquired in real-time and acquired from the intelligent manufacturing equipment into the state detection model, and confirm an operation state of the intelligent manufacturing equipment, where the operation state includes a fault state, an early warning state and a normal state, and specifically includes: the acquisition unit is used for respectively judging the preset number of decision trees of the real-time operation data input state detection model to acquire the preset number of fault states, early warning states and normal states; a first determining unit, configured to determine that an operation state of the intelligent manufacturing apparatus is a fault state when the fault state is the largest; the second determining unit is used for determining that the operation state of the intelligent manufacturing equipment is an early warning state when the early warning state is the most; and the third determining unit is used for determining that the operation state of the intelligent manufacturing equipment is a normal state when the normal state is the most.
The system of the embodiment of the application comprises: the intelligent manufacturing equipment comprises an acquisition module, a data acquisition module and a data analysis module, wherein the acquisition module is used for acquiring operation data of the intelligent manufacturing equipment and confirming a preset number of sampling samples based on the operation data, and the sampling samples comprise attribute characteristics of the preset number; the confirming module is used for confirming sample decision similarity coefficients corresponding to the target sampling period based on randomly calling a preset number of sampling samples and sampling samples of other sampling periods before the target sampling period in the target sampling period, wherein one sampling period corresponds to one decision tree; the computing module is used for confirming feature decision similarity coefficients corresponding to the target sampling period according to the preset number of attribute features of the sampling samples randomly fetched in the target sampling period; the judging module is used for judging whether a decision tree corresponding to the current target sampling period meets the establishment condition or not based on the sample decision similarity coefficient and the characteristic decision similarity coefficient so as to enter the condition of the next sampling period; the detection module is used for constructing a state detection model according to decision trees corresponding to a preset number of sampling periods so as to confirm the running state of the intelligent manufacturing equipment through the state detection model. And judging whether the decision tree corresponding to the target sampling period meets the establishment condition or not by acquiring the sample decision similarity coefficient and the feature decision similarity coefficient corresponding to the target sampling period so as to finally obtain a preset number of decision tree construction state detection models and confirm the running state of the intelligent manufacturing equipment, thereby improving the detection precision and further reducing the detection working cost.
Drawings
FIG. 1 is a block diagram of a digital twinning-based intelligent manufacturing facility data management system in accordance with an embodiment of the present application.
FIG. 2 is a first sub-block diagram of a digital twinning-based intelligent manufacturing apparatus data management system, in accordance with an embodiment of the present application.
FIG. 3 is a second sub-block diagram of a digital twinning-based intelligent manufacturing apparatus data management system, in accordance with an embodiment of the present application.
FIG. 4 is a third sub-block diagram of a digital twinning-based intelligent manufacturing apparatus data management system, in accordance with an embodiment of the present application.
FIG. 5 is a fourth sub-block diagram of a digital twinning-based intelligent manufacturing apparatus data management system, in accordance with an embodiment of the present application.
FIG. 6 is a fifth sub-block diagram of a digital twinning-based intelligent manufacturing apparatus data management system, in accordance with an embodiment of the present application.
FIG. 7 is a sixth sub-block diagram of a digital twinning-based intelligent manufacturing apparatus data management system, in accordance with an embodiment of the present application.
Detailed Description
In describing embodiments of the present application, words such as "exemplary," "or," "such as," and the like are used to mean serving as an example, instance, or illustration. Any embodiment or design described herein as "exemplary" or "for example" should not be construed as preferred or advantageous over other embodiments or designs. Rather, the use of words such as "exemplary," "or," "such as," and the like are intended to present related concepts in a concrete fashion.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used in the description of the application herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. It is to be understood that, unless otherwise indicated, a "/" means or. For example, A/B may represent A or B. The "and/or" in the present application is merely one association relationship describing the association object, indicating that three relationships may exist. For example, a and/or B may represent: a exists alone, A and B exist simultaneously, and B exists alone. "at least one" means one or more. "plurality" means two or more than two. For example, at least one of a, b or c may represent: seven cases of a, b, c, a and b, a and c, b and c, a, b and c.
It should be further noted that the terms "first" and "second" in the description and claims of the present application and the accompanying drawings are used for respectively similar objects, and are not used for describing a specific order or sequence. The method disclosed in the embodiments of the present application or the method shown in the flowchart, including one or more steps for implementing the method, may be performed in an order that the steps may be interchanged with one another, and some steps may be deleted without departing from the scope of the claims.
The embodiment of the application firstly provides a digital twinning-based intelligent manufacturing equipment data management system, which is applied to the field of digital data processing, and referring to fig. 1, the system comprises:
the system comprises an acquisition module 1, a storage module and a control module, wherein the acquisition module 1 is used for acquiring operation data of intelligent manufacturing equipment and confirming a preset number of sampling samples based on the operation data, wherein the sampling samples comprise attribute characteristics of the preset number.
The intelligent manufacturing equipment is manufacturing equipment with automation, intellectualization and networking characteristics, and achieves higher production efficiency, quality control and flexibility through advanced technologies such as integrated sensors, communication technology, data analysis and artificial intelligence. In this embodiment, the intelligent manufacturing device may be an intelligent device such as a numerically-controlled machine tool. The operation data refers to various data generated in the operation process of the equipment, and can be: sensor data: intelligent manufacturing facilities are typically equipped with various sensors for monitoring the operating state of the facility, environmental parameters, etc. The sensor data may include measurements of physical quantities such as temperature, pressure, humidity, vibration, etc.; production data: the intelligent manufacturing equipment can generate a large amount of production data in the production process, including production speed, yield, quality indexes and the like. These data can be used to monitor the operating state of the production line, to perform production efficiency analysis and quality control; fault data: the intelligent manufacturing equipment may have faults or abnormal conditions in the running process, and the fault data can record information such as fault type, occurrence time, duration time and the like of the equipment and is used for fault diagnosis and predictive maintenance; energy consumption data: the operation of the intelligent manufacturing equipment consumes energy, and the energy consumption data can record the energy consumption condition of the equipment and is used for energy management and optimization; running log: the operation log of the intelligent manufacturing equipment can record the operation time, operation record, alarm information and the like of the equipment and is used for tracing and analyzing the operation state of the equipment. These operational data may be collected and recorded by a data collection system built into the device or by an external data collection device. The collected data can be transmitted to a data center or a cloud platform through a network for storage and analysis, so that the applications of monitoring the running state of equipment, fault diagnosis, efficiency optimization and the like are realized. In this embodiment, the operation data is voltage data, current data, rotation speed data, vibration frequency data, temperature data and processing time data, and the data is collected by a voltage sensor, a current sensor, a rotation speed sensor, a vibration frequency sensor, a temperature sensor, a time recorder and other collection devices on the manufacturing equipment only.
It should be noted that, the operation data are voltage data, current data, rotation speed data, vibration frequency data, temperature data and processing time data, and the six data can represent the operation state of the intelligent manufacturing equipment, taking the intelligent manufacturing equipment as a numerical control machine tool as an example, when the numerical control machine tool equipment operates abnormally, the voltage and the current can be increased or decreased abnormally, the voltage and the current can be suddenly reduced due to the short circuit of the machine tool or due to the power supply, meanwhile, the numerical control machine tool is generally used as the cutting equipment, the temperature of machine tool parts is increased due to excessive wear or excessive cutting force of the machine tool in the operation process, the machine tool equipment is damaged due to the excessive temperature, and the excessive rotation speed can not only cause the excessive temperature but also cause the equipment damage due to the excessive rotation speed exceeding the limit value during the cutting of the numerical control machine tool. In addition, when the numerical control machine tool normally operates, the vibration frequency is stable, and abnormal vibration can be generated once the abnormal operation state occurs correspondingly; and secondly, under the abnormal operation state of the numerical control machine tool, the processing time of the numerical control machine tool is greatly different. Therefore, the six kinds of operation data are used as factors for judging the detection of the operation state of the intelligent manufacturing equipment, and the detection precision of the operation state can be further improved.
The sampling samples are used for constructing a subsequent decision tree, and each sampling sample can comprise 10 characteristic attributes, namely voltage data, current data, rotating speed data, vibration frequency data, temperature data and processing time data in the operation data, harmonic voltage frequency and amplitude and harmonic current frequency and amplitude. It should be noted that, the sequence length of each feature attribute is 600, and 10 feature attributes are acquired once for each acquisition period, i.e. 10 feature attributes are obtained in one acquisition period.
It should be noted that, the collected operation data needs to be subjected to data preprocessing, where the preprocessing may be a data cleaning algorithm, where the data cleaning algorithm is a part of data preprocessing, and is used to process problems of noise, missing values, abnormal values, and the like in the data, so as to improve quality and usability of the data. The working steps of the data cleaning algorithm can be summarized as follows: 1. noise treatment: noise is a random disturbance or error in the data that may adversely affect the analysis results. The goal of noise processing is to reduce or eliminate the effect of noise on the data. Common noise processing methods include smoothing, filtering, outlier removal, and the like. 2. Missing value processing: missing values are values for certain attributes or features in the data that are missing or unrecorded. The goal of the missing value processing is to fill in the missing values, leaving the data set intact. Common missing value processing methods include deleting samples of missing values, filling the missing values with means or median values, filling using regression or interpolation methods, and the like. 3. Outlier processing: the abnormal value is a value significantly different from other observed values, and may be caused by measurement errors, logging errors, data anomalies, or the like. The goal of outlier processing is to detect and process outliers to avoid interference with the data analysis. Common outlier processing methods include outlier detection based on statistical methods, outlier detection based on distance or density, outlier detection based on clustering or classification, and the like. 4. Data conversion: data conversion is the conversion of raw data into a form suitable for a particular algorithm or model. Common data conversion methods include normalization, discretization, feature selection, feature extraction, and the like. Data conversion may improve data interpretability, reduce dimensionality, reduce redundancy, and the like. 5. Data integration: data integration is the merging of multiple data sources or data sets into a whole. The goal of data integration is to eliminate duplicate data, resolve data conflicts, and consistency issues. Common data integration methods include merging, linking, correlation, and the like. 6. Data protocol: data reduction is to reduce the size of data by compression, sampling or aggregation. The goal of data reduction is to reduce the cost of storage and computation while maintaining important information of the data. Common data reduction methods include sampling, dimension reduction, clustering, and the like.
Further, the choice and application of the data cleansing algorithm depends on the specific data quality issue and analysis objective. Different problems and scenarios may require different data cleansing methods and techniques. Data cleansing is an important step in data analysis and machine learning, and is critical to obtaining accurate and reliable analysis results. The details of the pretreatment are not further limited, and the method is applicable by specifically referring to the steps.
And the confirmation module 2 is used for confirming sample decision similarity coefficients corresponding to the target sampling period based on randomly calling a preset number of sampling samples and sampling samples of other sampling periods before the target sampling period, wherein one sampling period corresponds to one decision tree.
The sampling period refers to that a part of samples are randomly extracted from original sampling samples, and part of characteristic attributes in the samples are randomly extracted to form a new training set for constructing a decision tree. The sample decision similarity coefficient refers to the similarity degree of a part of samples randomly extracted in the target sampling period and sampling samples randomly extracted in other sampling periods before the target sampling period, and when the sample decision similarity coefficient is larger, the decision tree corresponding to the target sampling period is proved to have larger similarity degree with the previous decision tree, and the classification result of the decision tree has lower referenceability.
It should be noted that, one sampling period is used to construct a decision tree, where the decision tree is used to classify data to obtain classification results, a preset number of sampling periods are generally used to obtain a preset number of decision trees, and the preset number of decision trees are used to predict the state of the intelligent manufacturing device to obtain a plurality of results, where the number of consistent results is the maximum as the final predicted result.
And the calculating module 3 is used for confirming the feature decision similarity coefficient corresponding to the target sampling period according to the attribute features of the preset number of the sampling samples randomly fetched in the target sampling period.
After obtaining the sampling samples randomly fetched in the target sampling period, randomly fetching a preset number of attribute features in the preset number of sampling samples to be used as the attribute features constructed by the decision tree, and simultaneously, calculating feature decision similarity coefficients corresponding to the target sampling period with the randomly fetched preset number of attribute features. The feature decision similarity coefficient refers to the similarity degree of the attribute features of the preset number of sampling samples randomly fetched in the target sampling period and the attribute features randomly fetched in other sampling periods before the target sampling period, when the feature decision similarity coefficient is larger, the fact that the similarity degree of a decision tree corresponding to the target sampling period and a previous decision tree is larger is proved, and the classification result of the decision tree is lower in referenceability.
It should be noted that, the number of randomly retrieving attribute features according to the randomly retrieved sample samples in the target sampling period may be preferably 4, and the corresponding data are ordered into a sequence, for example, the above 10 data attribute features: voltage data, current data, rotational speed data, vibration frequency data, temperature data, and time of machining data, and harmonic voltage frequency and amplitude and harmonic current frequency and amplitude, noted asThe random access attribute of the randomly accessed sampling sample in the target sampling period can be voltage data, harmonic voltage frequency, temperature data and current data, namely. Further, according to the random access of the attribute features of the sampling samples in the target sampling period, the feature sequence can be formed from large to small according to the gain of each feature information.
And the judging module 4 is used for judging whether the decision tree corresponding to the current target sampling period meets the establishment condition or not based on the sample decision similarity coefficient and the characteristic decision similarity coefficient so as to enter the next sampling period.
The method comprises the steps of obtaining a sample decision similarity coefficient through random sampling of a sampling sample, obtaining a feature decision similarity coefficient through random sampling of attribute features, judging whether the similarity degree of a decision tree corresponding to a current target sampling period and decision trees of other sampling periods before the target sampling period meets the establishment condition or not through the magnitudes of the sample decision similarity coefficient and the feature decision similarity coefficient, entering a next sampling period if the similarity degree meets the establishment condition, and continuing to establish the decision tree until a construction threshold of the decision tree is reached. If the current target sampling period is not met, judging that the decision tree corresponding to the current target sampling period does not meet the conditions, re-carrying out random sampling of the sampling sample and random sampling of the attribute characteristics, and continuously constructing the decision tree corresponding to the current target sampling period until the establishment conditions are met.
It should be noted that, by the sample decision similarity coefficient and the feature decision similarity coefficient, whether the decision tree established in each sampling period meets the establishment condition is judged, so that the situation that the similarity degree between the decision trees is high due to random sampling can be avoided, the accuracy in the subsequent classification judgment through the decision tree is high, and the working cost is reduced.
And the detection module 5 is used for constructing a state detection model according to decision trees corresponding to a preset number of sampling periods so as to confirm the running state of the intelligent manufacturing equipment through the state detection model.
After a preset number of decision trees are obtained through a preset number of sampling periods, a state detection model is built based on the preset number of decision trees, wherein the state detection model can be a random forest algorithm, and the random forest algorithm is an integrated learning method and is predicted by combining a plurality of decision trees. The algorithm steps of the random forest algorithm are as follows: 1. random sampling of data sets: a portion of the samples are randomly extracted from the original dataset to form a new training set. This process, called self-sampling, can make each training set have a certain variance. 2. Random selection of features: for the training process of each decision tree, a portion of the features from all of the features are randomly selected as inputs to the decision tree. This process, called random feature selection, may result in a different subset of features for each decision tree. 3. Building a decision tree: a decision tree is constructed using the selected feature subset and the self-sampled training set. The decision tree construction process may use common decision tree construction algorithms such as ID3, C4.5 or CART.4. Combination of multiple decision trees: and (3) repeating the step (2) and the step (3) to construct a plurality of decision trees. Each decision tree is constructed by randomly selecting features and randomly sampling a training set. 5. Integration of prediction results: for the classification problem, the random forest integrates the prediction results of each decision tree in a voting mode, and the class with the largest vote is selected as the final prediction result. For regression problem, the random forest integrates the prediction result of each decision tree in an average value mode. It should be noted that, the random forest algorithm is adopted as the detection model, which can include the following advantages: high-dimensional data and large-scale data sets can be processed; the method has good robustness to abnormal values and noise; the importance of the feature can be assessed; the parallel processing can be realized, and the training and prediction speeds are increased.
It should be noted that the decision tree construction algorithm is a supervised learning algorithm based on a tree structure, and is used for solving the classification and regression problems. Decision trees construct a tree by recursively halving the dataset, where each non-leaf node represents a feature and each leaf node represents a class or a value. The basic principle of the decision tree construction algorithm is as follows: 1. feature selection: one feature is selected from the training set as the partitioning feature for the current node. Common feature selection methods include information gain, information gain ratio, base index, and the like. Selecting the appropriate features may make the partitioned subsets purer, i.e. the samples of the same class are more concentrated. 2. Dividing the data set: the data set is divided into a plurality of subsets according to the selected features. Each subset corresponds to a branch of the current node. The division may be binary or multiple. 3. Recursively constructing a subtree: for each subset, steps 1 and 2 are repeated, recursively building subtrees. Until a termination condition is met, such as reaching a maximum depth, the samples in the subset all belonging to the same class, or the number of samples in the subset is less than a certain threshold. 4. Pruning: in order to avoid overfitting, pruning can be performed on the constructed decision tree. The goal of pruning is to improve generalization capability by reducing the complexity of the decision tree. 5. Prediction and classification: and predicting and classifying by using the constructed decision tree. For classification problems, the predicted class is obtained by traversing from the root node, along the branches of the tree, according to the values of the features, and finally reaching the leaf nodes. For regression problems, the average or other statistics of leaf nodes may be used as the prediction result. The specific construction process is substituted by itself with reference to the above, and the present disclosure is not further limited and described.
The system of the embodiment of the application comprises: the intelligent manufacturing equipment comprises an acquisition module, a data acquisition module and a data analysis module, wherein the acquisition module is used for acquiring operation data of the intelligent manufacturing equipment and confirming a preset number of sampling samples based on the operation data, and the sampling samples comprise attribute characteristics of the preset number; the confirming module is used for confirming sample decision similarity coefficients corresponding to the target sampling period based on randomly calling a preset number of sampling samples and sampling samples of other sampling periods before the target sampling period in the target sampling period, wherein one sampling period corresponds to one decision tree; the computing module is used for confirming feature decision similarity coefficients corresponding to the target sampling period according to the preset number of attribute features of the sampling samples randomly fetched in the target sampling period; the judging module is used for judging whether a decision tree corresponding to the current target sampling period meets the establishment condition or not based on the sample decision similarity coefficient and the characteristic decision similarity coefficient so as to enter the condition of the next sampling period; the detection module is used for constructing a state detection model according to decision trees corresponding to a preset number of sampling periods so as to confirm the running state of the intelligent manufacturing equipment through the state detection model. And judging whether the decision tree corresponding to the target sampling period meets the establishment condition or not by acquiring the sample decision similarity coefficient and the feature decision similarity coefficient corresponding to the target sampling period so as to finally obtain a preset number of decision tree construction state detection models and confirm the running state of the intelligent manufacturing equipment, thereby improving the detection precision and further reducing the detection working cost.
In an embodiment of the present application, and referring to fig. 2, the confirmation module 2 is configured to confirm, in a target sampling period, a sample decision similarity coefficient corresponding to the target sampling period based on randomly calling a preset number of sampling samples and sampling samples of other sampling periods before the target sampling period, where one sampling period corresponds to one decision tree, and specifically includes:
the first sub-confirmation module 21 is configured to randomly call a preset number of sampling samples in a target sampling period and sequence the repetition frequency of sampling samples in other sampling periods before the target sampling period, so as to construct a target sampling sequence corresponding to the target sampling period.
After sampling samples of a preset number of sampling samples and sampling samples of other sampling periods before a target sampling period are randomly acquired in the target sampling period, counting the repeated occurrence frequency of each sampling sample, forming a statistical histogram according to the repeated occurrence frequency of each sampling sample (namely, the abscissa is the sampling sample and the ordinate is the occurrence frequency), sequencing the abscissa samples from high to low according to the occurrence frequency, and constructing a target sampling sequence corresponding to the target sampling period.
The second sub-confirmation module 22 is configured to calculate a sample decision similarity coefficient corresponding to the target sampling period based on the maximum repetition frequency of the target sampling sequence, the number of sampling samples, and the attribute feature sequence corresponding to each sampling sample.
After the target sampling sequence is obtained, the maximum repetition frequency, the number of sampling samples and the attribute feature sequence corresponding to each sampling sample of the target sampling sequence can be confirmed based on the target sampling sequence, and the attribute feature sequence is used as a calculation parameter and is input into a preset calculation algorithm to calculate a sample decision similarity coefficient corresponding to a target sampling period.
Specifically, the second sub-confirmation module 22 is configured to calculate, based on the maximum repetition frequency of the target sampling sequence, the number of sampling samples, and the attribute feature sequence corresponding to each sampling sample, a sample decision similarity coefficient corresponding to the target sampling period, and specifically includes:
wherein,is the firstSample decision similarity coefficients corresponding to the individual target sampling periods,is the firstThe maximum repetition frequency of the target sample sequence corresponding to the target sample period,is the firstThe number of sample samples of the target sample sequence corresponding to the target sample period, Is the firstThe first target sampling sequence corresponding to each target sampling periodThe sequence of attribute features corresponding to the individual sample samples,is the firstThe first target sampling sequence corresponding to each target sampling periodThe sequence of attribute features corresponding to the individual sample samples,to calculate the edit distance between two attribute feature sequences.
It should be noted that if there are samples with a larger repetition frequency in the target sampling period, i.e., the value of the maximum repetition frequency is larger, thenIf the difference between the attribute feature sequences corresponding to the sampling samples in the target sampling period is larger, the attribute feature sequences corresponding to the target sampling periodThe larger the editing distance between the two sampling periods is, the larger the sample decision similarity coefficient corresponding to the target sampling period is.
It should be noted that the edit distance is an index for measuring the similarity between two character strings. It measures the minimum number of editing operations required to convert one string to another. Editing operations include three operations of insertion, deletion, and replacement. Through these operations, one character in one character string can be converted into another character, or one character can be inserted or deleted in the character string. The edit distance may be calculated using different algorithms, with the most common algorithms being the Levenshtein distance and the Damerau-Levenshtein distance. Both algorithms are based on the idea of dynamic planning, by constructing a two-dimensional matrix to calculate the edit distance. The specific calculation mode is not further limited, and the method can be realized by the contents recorded in the prior art.
In one embodiment of the present application, referring to fig. 3, the calculating module 3 is configured to determine, according to a preset number of attribute features of a sample randomly fetched in a target sampling period, a feature decision similarity coefficient corresponding to the target sampling period, and specifically includes:
the first sub-calculation module 31 is configured to confirm a sampling attribute feature sequence corresponding to the target sampling period according to a preset number of attribute features of sampling samples randomly fetched in the target sampling period.
After the preset number of attribute features of the sampling samples randomly called in the target sampling period are obtained, the preset number of attribute features are randomly obtained to confirm a sampling attribute feature sequence corresponding to the target sampling period. For example, a predetermined number of attribute features of a sample randomly fetched in a target sample period includes a predetermined number of 10 data attribute features: voltage data, current data, rotational speed data, vibration frequency data, temperature data, and time of machining data, and harmonic voltage frequency and amplitude and harmonic current frequency and amplitude, noted asOrder of thenThe characteristic sequence of the sampling attribute corresponding to the standard sampling period is voltage data, harmonic voltage frequency, temperature data and current data, namely 。
The second sub-calculation module 32 is configured to construct a target attribute feature matrix corresponding to the target sampling period based on the target attribute feature sequence corresponding to the target sampling period and other attribute feature sequences corresponding to other sampling periods before the target sampling period.
Specifically, referring to fig. 4, the second sub-calculating module 32 is configured to construct a target attribute feature matrix corresponding to the target sampling period based on the target attribute feature sequence corresponding to the target sampling period and other attribute feature sequences corresponding to other sampling periods before the target sampling period, and specifically includes:
a construction unit 321, configured to sort other attribute feature sequences corresponding to other sampling periods before the target sampling period according to columns, and construct other attribute feature matrices corresponding to other sampling periods;
and the confirmation unit 322 is configured to confirm the target attribute feature matrix corresponding to the target sampling period by performing a comparison process of a preset rule on the target attribute feature sequence corresponding to the target sampling period and column vectors of other attribute feature matrices corresponding to other sampling periods.
After the target attribute feature sequences corresponding to the target sampling periods are obtained, other attribute feature sequences corresponding to other sampling periods before the target sampling periods are ordered according to the sequence order to form other attribute feature matrices, wherein the formats of the attribute feature sequences corresponding to each sampling period are the same, the attribute features in each sequence are sequentially arranged according to the information gain of the attribute feature sequences, and the attribute feature sequences are arranged according to the sequence order, namely, each attribute feature sequence in the other attribute feature sequences is used as a column and ordered according to the sequence order, so that one other attribute feature matrix is formed, for example, the following formula shows:
Then, the target attribute feature sequences corresponding to the target sampling periods are integrated into the other attribute feature matrixesIn the method, the attribute features of each row are compared in pairs, the attribute features are set to be 1 if the attribute features are the same, and the attribute features are set to be 0 if the attribute features are different, so as to obtain a target attribute feature matrix corresponding to the target sampling period, for example, the target attribute feature matrix:
And the third sub-calculation module 33 is configured to perform a preset conversion algorithm on the target attribute feature matrix, and confirm a target attribute feature sequence corresponding to the target sampling period.
Wherein, the analysis and explanation are carried out by combining the above examples, the target attribute feature matrixThe target attribute feature matrix is a matrix in binary formEach column of elements in the sequence form binary codes, decimal marks are converted into data points, and the target attribute characteristic sequence corresponding to the target sampling period is obtained.
And a fourth sub-calculation module 34, configured to calculate a feature decision similarity coefficient corresponding to the target sampling period based on the number of data points in the target attribute feature sequence and the data value corresponding to each data point.
The fourth sub-calculation module is configured to calculate, based on the number of data points in the target attribute feature sequence and a data value corresponding to each data point, a feature decision similarity coefficient corresponding to a target sampling period, and specifically includes:
Wherein,is the firstThe feature decision similarity coefficients corresponding to the individual target sampling periods,is the firstThe number of data points in the target attribute feature sequence corresponding to the target sampling period,is the firstThe first target attribute feature sequence corresponding to each target sampling periodData values for data points.
In one embodiment of the present application, referring to fig. 5, the determining module 4 is configured to determine, based on the sample decision similarity coefficient and the feature decision similarity coefficient, whether a decision tree corresponding to a current target sampling period meets a set-up condition, so as to enter a next sampling period condition, and specifically includes:
the first sub-judging module 41 is configured to input the sample decision similarity coefficient and the feature decision similarity coefficient into a preset decision similarity calculation formula, and calculate the decision similarity of the decision tree corresponding to the target sampling period.
And the second sub-judging module 42 is configured to determine that the target sampling period needs to be resampled when the decision similarity of the decision tree corresponding to the target sampling period is greater than a preset similarity threshold.
And the third sub-judging module 43 is configured to determine that the decision tree corresponding to the target sampling period satisfies a condition when the decision similarity of the decision tree corresponding to the target sampling period is less than or equal to a preset similarity threshold, and enter a condition of a next sampling period.
Specifically, the first sub-determining module 41 is configured to input the sample decision similarity coefficient and the feature decision similarity coefficient into a preset decision similarity calculation formula, and calculate the decision similarity of the decision tree corresponding to the target sampling period, and specifically includes:
wherein,is the firstSample decision similarity coefficients corresponding to the individual target sampling periods,is the firstSample decision similarity coefficients corresponding to the individual target sampling periods,is a normalization function.
In one embodiment of the present application, referring to fig. 6, the detecting module 5 is configured to construct a state detection model according to a decision tree corresponding to a preset number of sampling periods, so as to confirm, by using the state detection model, an operation state of the intelligent manufacturing apparatus, and specifically includes:
the first sub-detection module 51 is configured to construct a state detection model according to decision trees corresponding to a preset number of sampling periods.
Specifically, the number of decision trees is preferably 100, the state detection model may be a random forest algorithm, and specific working steps of the random forest algorithm are described in the foregoing embodiments and are not described herein.
The second sub-detection module 52 is configured to input real-time operation data acquired in real-time and acquired from the intelligent manufacturing apparatus into the state detection model, and confirm an operation state of the intelligent manufacturing apparatus, where the operation state includes a fault state, an early warning state, and a normal state.
Specifically, referring to fig. 7, the second sub-detection module 52 is configured to input real-time operation data acquired in real-time and acquired from the intelligent manufacturing apparatus into the state detection model, and confirm an operation state of the intelligent manufacturing apparatus, where the operation state includes a fault state, an early warning state, and a normal state, and specifically includes:
the obtaining unit 521 is configured to determine a preset number of decision trees of the real-time operation data input state detection model, and obtain a preset number of fault states, early warning states and normal states;
a first determining unit 522, configured to determine that the operation state of the smart manufacturing device is a fault state when the fault state is the largest;
a second determining unit 523, configured to determine, when the early warning state is the largest, that the operation state of the smart manufacturing device is an early warning state;
and a third determining unit 524, configured to determine that the operation state of the smart manufacturing device is a normal state when the normal state is the maximum.
The system of the embodiment of the application comprises: the intelligent manufacturing equipment comprises an acquisition module, a data acquisition module and a data analysis module, wherein the acquisition module is used for acquiring operation data of the intelligent manufacturing equipment and confirming a preset number of sampling samples based on the operation data, and the sampling samples comprise attribute characteristics of the preset number; the confirming module is used for confirming sample decision similarity coefficients corresponding to the target sampling period based on randomly calling a preset number of sampling samples and sampling samples of other sampling periods before the target sampling period in the target sampling period, wherein one sampling period corresponds to one decision tree; the computing module is used for confirming feature decision similarity coefficients corresponding to the target sampling period according to the preset number of attribute features of the sampling samples randomly fetched in the target sampling period; the judging module is used for judging whether a decision tree corresponding to the current target sampling period meets the establishment condition or not based on the sample decision similarity coefficient and the characteristic decision similarity coefficient so as to enter the condition of the next sampling period; the detection module is used for constructing a state detection model according to decision trees corresponding to a preset number of sampling periods so as to confirm the running state of the intelligent manufacturing equipment through the state detection model. And judging whether the decision tree corresponding to the target sampling period meets the establishment condition or not by acquiring the sample decision similarity coefficient and the feature decision similarity coefficient corresponding to the target sampling period so as to finally obtain a preset number of decision tree construction state detection models and confirm the running state of the intelligent manufacturing equipment, thereby improving the detection precision and further reducing the detection working cost.
The flowcharts and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to embodiments of the present disclosure. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). In some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. In the description corresponding to the flowcharts and block diagrams in the figures, operations or steps corresponding to different blocks may also occur in different orders than that disclosed in the description, and sometimes no specific order exists between different operations or steps. For example, two consecutive operations or steps may actually be performed substantially in parallel, they may sometimes be performed in reverse order, which may be dependent on the functions involved. Each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
It will be evident to those skilled in the art that the application is not limited to the details of the foregoing illustrative embodiments, and that the present application may be embodied in other specific forms without departing from the spirit or essential characteristics thereof. The above-described embodiments of the application are, therefore, to be considered in all respects as illustrative and not restrictive, the scope of the application being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein.
Claims (7)
1. An intelligent manufacturing equipment data management system based on digital twinning is applied to the field of digital data processing, and is characterized in that the system comprises:
the intelligent manufacturing equipment comprises an acquisition module, a data acquisition module and a data analysis module, wherein the acquisition module is used for acquiring operation data of the intelligent manufacturing equipment and confirming a preset number of sampling samples based on the operation data, and the sampling samples comprise attribute characteristics of the preset number;
the confirming module is used for confirming sample decision similarity coefficients corresponding to the target sampling period based on randomly calling a preset number of sampling samples and sampling samples of other sampling periods before the target sampling period in the target sampling period, wherein one sampling period corresponds to one decision tree;
The computing module is used for confirming feature decision similarity coefficients corresponding to the target sampling period according to the preset number of attribute features of the sampling samples randomly fetched in the target sampling period;
the judging module is used for judging whether a decision tree corresponding to the current target sampling period meets the establishment condition or not based on the sample decision similarity coefficient and the characteristic decision similarity coefficient so as to enter the next sampling period;
the detection module is used for constructing a state detection model according to decision trees corresponding to a preset number of sampling periods so as to confirm the running state of the intelligent manufacturing equipment through the state detection model;
the confirmation module is configured to confirm, in a target sampling period, a sample decision similarity coefficient corresponding to the target sampling period based on randomly calling a preset number of sampling samples and sampling samples of other sampling periods before the target sampling period, where one sampling period corresponds to one decision tree, and specifically includes:
the first sub-confirmation module is used for randomly calling a preset number of sampling samples in a target sampling period and carrying out size sorting on the repetition frequency of the sampling samples in other sampling periods before the target sampling period to construct a target sampling sequence corresponding to the target sampling period;
The second sub-confirmation module is used for calculating sample decision similarity coefficients corresponding to the target sampling period based on the maximum repetition frequency of the target sampling sequence, the number of sampling samples and the attribute feature sequence corresponding to each sampling sample;
the second sub-confirmation module is configured to calculate a sample decision similarity coefficient corresponding to a target sampling period based on a maximum repetition frequency of the target sampling sequence, the number of sampling samples, and an attribute feature sequence corresponding to each sampling sample, and specifically includes:
wherein,is->Sample decision similarity coefficient corresponding to each target sampling period, < ->Is->Target sampling corresponding to each target sampling periodMaximum repetition rate of the sequence,/->Is->The number of sample samples of the target sample sequence corresponding to the target sample period, is->Is->The first +.of the target sampling sequence corresponding to the target sampling period>The sequence of attribute features corresponding to the individual sample samples,is->The first +.of the target sampling sequence corresponding to the target sampling period>The sequence of attribute features corresponding to the individual sample samples,calculating an editing distance between two attribute feature sequences;
the computing module is configured to determine, according to a preset number of attribute features of sampling samples randomly fetched in a target sampling period, feature decision similarity coefficients corresponding to the target sampling period, and specifically includes:
The first sub-calculation module is used for confirming a sampling attribute feature sequence corresponding to the target sampling period according to the attribute features of the preset number of sampling samples randomly fetched in the target sampling period;
the second sub-calculation module is used for constructing a target attribute feature matrix corresponding to the target sampling period based on the target attribute feature sequence corresponding to the target sampling period and other attribute feature sequences corresponding to other sampling periods before the target sampling period;
the third sub-calculation module is used for carrying out preset conversion algorithm processing on the target attribute feature matrix and confirming a target attribute feature sequence corresponding to a target sampling period;
and the fourth sub-calculation module is used for calculating the feature decision similarity coefficient corresponding to the target sampling period based on the number of data points in the target attribute feature sequence and the data value corresponding to each data point.
2. The digital twinning-based intelligent manufacturing equipment data management system according to claim 1, wherein the second sub-calculation module is configured to construct a target attribute feature matrix corresponding to a target sampling period based on a target attribute feature sequence corresponding to the target sampling period and other attribute feature sequences corresponding to other sampling periods before the target sampling period, and specifically includes:
The construction unit is used for sequencing other attribute feature sequences corresponding to other sampling periods before the target sampling period according to the columns, and constructing other attribute feature matrixes corresponding to the other sampling periods;
and the confirmation unit is used for carrying out comparison processing of preset rules through the column vectors of the target attribute feature sequences corresponding to the target sampling periods and other attribute feature matrices corresponding to other sampling periods, and confirming the target attribute feature matrices corresponding to the target sampling periods.
3. The digital twinning-based intelligent manufacturing equipment data management system according to claim 2, wherein the fourth sub-calculation module is configured to calculate a feature decision similarity coefficient corresponding to a target sampling period based on a number of data points in the target attribute feature sequence and a data value corresponding to each data point, and specifically includes:
wherein,is->Characteristic decision similarity coefficient corresponding to each target sampling period, < ->Is->The number of data points in the target attribute characteristic sequence corresponding to each target sampling period is +.>Is->The first +.in the target attribute feature sequence corresponding to the target sampling period>Data values for data points.
4. The digital twinning-based intelligent manufacturing equipment data management system according to claim 3, wherein the judging module is configured to judge, based on the sample decision similarity coefficient and the feature decision similarity coefficient, whether a decision tree corresponding to a current target sampling period meets an establishment condition, so as to enter a next sampling period condition, and specifically includes:
The first sub-judging module is used for inputting the sample decision similarity coefficient and the characteristic decision similarity coefficient into a preset decision similarity calculation formula and calculating the decision similarity of a decision tree corresponding to the target sampling period;
the second sub-judging module is used for determining that the target sampling period needs to be resampled when the decision similarity of the decision tree corresponding to the target sampling period is larger than a preset similarity threshold;
and the third sub-judging module is used for determining that the decision tree corresponding to the target sampling period meets the condition when the decision similarity of the decision tree corresponding to the target sampling period is smaller than or equal to a preset similarity threshold value, and entering the condition of the next sampling period.
5. The digital twin based intelligent manufacturing equipment data management system according to claim 4, wherein the first sub-judging module is configured to input the sample decision similarity coefficient and the feature decision similarity coefficient into a preset decision similarity calculation formula, and calculate a decision similarity of a decision tree corresponding to the target sampling period, and specifically includes:
wherein,is->Sample decision similarity coefficient corresponding to each target sampling period, < ->Is->Sample decision similarity coefficient corresponding to each target sampling period, < - >Is a normalization function.
6. The digital twinning-based intelligent manufacturing apparatus data management system according to any one of claims 1 to 5, wherein the detection module is configured to construct a state detection model according to a decision tree corresponding to a preset number of sampling periods, so as to confirm an operation state of the intelligent manufacturing apparatus through the state detection model, and specifically includes:
the first sub-detection module is used for constructing a state detection model according to decision trees corresponding to a preset number of sampling periods;
the second sub-detection module is used for inputting real-time operation data acquired in real time and used for acquiring the intelligent manufacturing equipment into the state detection model and confirming the operation state of the intelligent manufacturing equipment, wherein the operation state comprises a fault state, an early warning state and a normal state.
7. The digital twinning-based intelligent manufacturing apparatus data management system according to claim 6, wherein the second sub-detection module is configured to input real-time operation data acquired in real-time for collecting intelligent manufacturing apparatuses into the state detection model, and confirm an operation state of the intelligent manufacturing apparatuses, where the operation state includes a fault state, an early warning state, and a normal state, and specifically includes:
The acquisition unit is used for respectively judging the preset number of decision trees of the real-time operation data input state detection model to acquire the preset number of fault states, early warning states and normal states;
a first determining unit, configured to determine that an operation state of the intelligent manufacturing apparatus is a fault state when the fault state is the largest;
the second determining unit is used for determining that the operation state of the intelligent manufacturing equipment is an early warning state when the early warning state is the most;
and the third determining unit is used for determining that the operation state of the intelligent manufacturing equipment is a normal state when the normal state is the most.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202311134407.9A CN116861204B (en) | 2023-09-05 | 2023-09-05 | Intelligent manufacturing equipment data management system based on digital twinning |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202311134407.9A CN116861204B (en) | 2023-09-05 | 2023-09-05 | Intelligent manufacturing equipment data management system based on digital twinning |
Publications (2)
Publication Number | Publication Date |
---|---|
CN116861204A CN116861204A (en) | 2023-10-10 |
CN116861204B true CN116861204B (en) | 2023-12-08 |
Family
ID=88236312
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202311134407.9A Active CN116861204B (en) | 2023-09-05 | 2023-09-05 | Intelligent manufacturing equipment data management system based on digital twinning |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN116861204B (en) |
Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP3696622A1 (en) * | 2019-02-14 | 2020-08-19 | Rockwell Automation Technologies, Inc. | Ai extensions and intelligent model validation for an industrial digital twin |
CN113516297A (en) * | 2021-05-26 | 2021-10-19 | 平安国际智慧城市科技股份有限公司 | Prediction method and device based on decision tree model and computer equipment |
WO2022257925A1 (en) * | 2021-06-09 | 2022-12-15 | 五邑大学 | Fault prediction method and apparatus based on digital twin, server, and storage medium |
CN115659177A (en) * | 2022-10-25 | 2023-01-31 | 招联消费金融有限公司 | Method and device for generating data recommendation model and computer equipment |
CN115718466A (en) * | 2022-11-21 | 2023-02-28 | 广东工业大学 | Digital twin workshop fault prediction method based on random forest and analytic hierarchy process |
CN116048958A (en) * | 2022-11-17 | 2023-05-02 | 中山大学 | Medical robot control software test data generation method and injection method |
CN116070802A (en) * | 2023-04-03 | 2023-05-05 | 中国电子科技集团公司第十五研究所 | Intelligent monitoring operation and maintenance method and system based on data twinning |
CN116501005A (en) * | 2023-06-30 | 2023-07-28 | 广州力控元海信息科技有限公司 | Digital twin linkage factory operation management method and system |
CN116561681A (en) * | 2023-05-16 | 2023-08-08 | 西安电子科技大学 | Digital twinning-based fusion type fault diagnosis method |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US11348048B2 (en) * | 2018-02-23 | 2022-05-31 | Tata Consultancy Services Limited | Method and system for pre-emptive deviation detection for telecom process flow using digital twin |
-
2023
- 2023-09-05 CN CN202311134407.9A patent/CN116861204B/en active Active
Patent Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP3696622A1 (en) * | 2019-02-14 | 2020-08-19 | Rockwell Automation Technologies, Inc. | Ai extensions and intelligent model validation for an industrial digital twin |
CN113516297A (en) * | 2021-05-26 | 2021-10-19 | 平安国际智慧城市科技股份有限公司 | Prediction method and device based on decision tree model and computer equipment |
WO2022257925A1 (en) * | 2021-06-09 | 2022-12-15 | 五邑大学 | Fault prediction method and apparatus based on digital twin, server, and storage medium |
CN115659177A (en) * | 2022-10-25 | 2023-01-31 | 招联消费金融有限公司 | Method and device for generating data recommendation model and computer equipment |
CN116048958A (en) * | 2022-11-17 | 2023-05-02 | 中山大学 | Medical robot control software test data generation method and injection method |
CN115718466A (en) * | 2022-11-21 | 2023-02-28 | 广东工业大学 | Digital twin workshop fault prediction method based on random forest and analytic hierarchy process |
CN116070802A (en) * | 2023-04-03 | 2023-05-05 | 中国电子科技集团公司第十五研究所 | Intelligent monitoring operation and maintenance method and system based on data twinning |
CN116561681A (en) * | 2023-05-16 | 2023-08-08 | 西安电子科技大学 | Digital twinning-based fusion type fault diagnosis method |
CN116501005A (en) * | 2023-06-30 | 2023-07-28 | 广州力控元海信息科技有限公司 | Digital twin linkage factory operation management method and system |
Non-Patent Citations (2)
Title |
---|
基于数字孪生的泛在电力物联网模型研究;纪志伟;;电力学报(03);全文 * |
基于数字孪生的船舶预测性维护;李福兴;李璐爔;彭友;;船舶工程(S1);全文 * |
Also Published As
Publication number | Publication date |
---|---|
CN116861204A (en) | 2023-10-10 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN115578015B (en) | Sewage treatment whole process supervision method, system and storage medium based on Internet of things | |
CN110895526A (en) | Method for correcting data abnormity in atmosphere monitoring system | |
CN111340063B (en) | Data anomaly detection method for coal mill | |
CN111027615B (en) | Middleware fault early warning method and system based on machine learning | |
CN113569903A (en) | Method, system, equipment, medium and terminal for predicting abrasion of numerical control machine tool cutter | |
CN112905380A (en) | System anomaly detection method based on automatic monitoring log | |
CN114118224A (en) | Neural network-based system-wide remote measurement parameter anomaly detection system | |
CN114399066B (en) | Mechanical equipment predictability maintenance system and maintenance method based on weak supervision learning | |
CN116914917A (en) | Big data-based monitoring and management system for operation state of power distribution cabinet | |
CN117421994A (en) | Edge application health monitoring method and system | |
CN110399278B (en) | Alarm fusion system and method based on data center anomaly monitoring | |
CN116599857A (en) | Digital twin application system suitable for multiple scenes of Internet of things | |
CN115034278A (en) | Performance index abnormality detection method and device, electronic equipment and storage medium | |
CN113379033A (en) | Cable health state intelligent early warning method based on time-space network enhanced deep learning | |
CN114326593B (en) | Cutter life prediction system and method | |
CN116167370A (en) | Log space-time characteristic analysis-based distributed system anomaly detection method | |
CN109299201B (en) | Power plant production subsystem abnormity monitoring method and device based on two-stage clustering | |
CN117473351B (en) | Power supply information remote transmission system based on Internet of things | |
CN117827928A (en) | Database inspection method based on abnormal feature extraction | |
WO2024027487A1 (en) | Health degree evaluation method and apparatus based on intelligent operations and maintenance scene | |
CN116861204B (en) | Intelligent manufacturing equipment data management system based on digital twinning | |
CN116862109A (en) | Regional carbon emission situation awareness early warning method | |
CN116170200A (en) | Power monitoring system time sequence abnormality detection method, system, equipment and storage medium | |
CN116720079A (en) | Wind driven generator fault mode identification method and system based on multi-feature fusion | |
CN116628609A (en) | Intelligent monitoring method and system for multi-source data analysis |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |